Project Summary/Abstract DNA cytosine methylation (5-methylcytosine or 5mC) is the main epigenetic mechanism in human gene expression regulation. This methylation is oxidized by the human TET family enzymes to 5-hydroxymethylcytosine (5hmC) in an active demethylation process. While 5mC is a mark for gene repression in general and is also used to suppress repetitive elements in the human genome, recent studies showed that 5hmC tends to mark active loci, as gene activation requires removal of 5mC in gene bodies, promoters, and enhancers. The presence of 5hmC indicates open chromatin loci, whereas heavy 5mC methylation often marks heterochromatin regions. Therefore, genomic locations of 5hmC and 5mC provide genome-wide, comprehensive information on chromatin activation and repression, which are suitable marks for classifying individual human cells. The locations of 5hmC and 5mC can also store information of gene activation in frozen or fragmented genomic samples, such as DNA from biopsy samples or cell- free DNA (cfDNA), providing additional advantages in clinical diagnosis and prognosis of various human diseases. Building on our previous successes in inventing enabling technologies that label and sequence 5hmC and 5mC, we have recently developed robust procedures to map 5hmC and related cytosine modifications genome-wide using 1,000 cells. In this application we propose new approaches that can map both 5hmC and 5hmC at base-resolution down to the single-cell level. We plan to test and validate these new methods by investigating cell-cell heterogeneity using Tet2-mutant acute myeloid leukemia stem cells versus controls. We will also validate the new methods to map 5hmC and 5mC in other systems with limited DNA. In addition, we propose a new and highly useful method for single-cell bisulfite sequencing of 5mC in DNA by pre-amplify the limited DNA samples to much higher abundances compatible with most downstream analysis methods, with 5mC sites replicated faithfully during the amplification. The proposed research will provide urgently needed tools for the PI's group and the broader scientific community to study a range of questions in biomedical research and clinical biomarker discoveries using limited input DNA and at the single-cell level.